Basic research on articular cartilage biochemistry and metabolism has identified compounds (e.g., metalloproteinase inhibitors) that have shown great potential in animal models as disease-modifying drugs for osteoarthritis (DMOADs). The osteoarthritis (OA) research community is in general agreement that narrowing of the joint space (JSN), measured in serial radiographs, is a surrogate for loss of articular cartilage that can, with standardized positioning of the subject be measured with sufficient accuracy to permit detection of a DMOAD effect. However, research on the effects of DMOADs has been hampered by inconsistent and incomplete information on which to base decisions regarding the most appropriate target population(s) and primary outcome variables (i.e., measures reflecting progression f OA). It has been suggested that the presence of Heberden's nodes or results of bone scintigraphy may help identify subjects at increased risk for progressive OA and that recent epidemiologic data suggest that a select population of persons with radiographic evidence of unilateral knee OA and the cardinal risk factors for bilateral disease (female sex, obesity, middle age) may afford an opportunity to observe (and prevent) the rapid onset of OA in a joint at high risk for OA - i.e., the radiographically normal contralateral knee. However, the degree to which bony changes of OA in the high-risk contralateral knee of this select group (i.e. osteophytosis) are accompanied by JSN is unknown. The study proposed herein will provide information which will be useful in the design of future DMOAD trials; 150 subjects will be enrolled in a longitudinal study of OA progression: 60 women age 45-64 with radiographic evidence of unilateral knee OA, 30 men age 45-64 with bilateral knee OA, 30 women and 30 men age 65-74 with bilateral knee OA. Because obesity is a recognized risk factor for knee OA, all subjects will have body mass index (BMI) in the upper tertile for sex- , race-, and age-adjusted norms. Knee OA will be mild/moderate at baseline [osteophytes and greater than or equal to 2 mm of minimum medial tibiofemoral joint space width (JSW)]. Subjects will undergo x-ray examination of knees and hands and algofunctional assessment (WOMAC) at baseline. Those with unilateral knee OA will also undergo baseline bone scintigraphy, results of which will be related to x-ray changes over time to ascertain the predictive value of scintigraphy for progression of knee OA. Knee radiography will be repeated 15 and 30 months after baseline. Bony changes of knee OA (e.g., osteophytes, sclerosis) will be graded semiquantitatively while, for precision and generalizability to a future DMOAD trial, JSN will be quantitated from serial computerized measurements of minimum medial compartment of JSW in digitized images of both knees obtained with standardized positioning under fluoroscopy, with correction for image magnification.